SemaPLC Project-Context Track
Agents implement plant control logic inside existing Structured Text projects. The evaluator measures compilation, static checks, and agreement with reference outputs during execution.
The study compares seven model backends on 65 adapted Spec2Control sections. The public release includes static checks and reference projects. It omits the runtime evaluator.
- Made by
- Midea AIRC and collaborators
- Released
- August 2026
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal
- Items from
- Another benchmark
- Refresh
- One fixed set
- Run by
- A host
- Based on
- Spec2Control
- In views
- Agents
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- June 2026
- Scores refreshed
- November 20th, 2026
GPT-5.5 leads with 100% among models released by June 2026
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The headline is the integrated compilation rate, the first of three separate measures in Table II. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in June 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 GPT-5.5 100%
- 2
DeepSeek-V4-Pro 95.4%
- 3
MiniMax-M3 95.4%
- 4 GLM-5.2 95.4%
- 5
DeepSeek-V4-Flash 84.6%
- 6
MiniMax-M2.7 81.5%
- 7
Qwen3.5-Plus 80%
Source: The SemaPLC Project-Context Track paper
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 100% | LLM4PLC | Apr 2026 | Official leaderboard |
| 2 | | 95.4% | AutoPLC | Apr 2026 | Official leaderboard |
| 3 | | 95.4% | AutoPLC | Jun 2026 | Official leaderboard |
| 4 | GLM-5.2 | 95.4% | AutoPLC | Jun 2026 | Official leaderboard |
| 5 | | 84.6% | SemaPLC | Apr 2026 | Official leaderboard |
| 6 | | 81.5% | SemaPLC | Mar 2026 | Official leaderboard |
| 7 | | 80% | SemaPLC | Feb 2026 | Official leaderboard |
Rankings by part
We read 2 parts from the benchmark’s paper: Static behavior and Dynamic behavior. Each ranks models on one part of the test. GPT-5.5 leads both. Arrows show how far a model moved from its overall rank among the models we track.
Static behavior
7 models we track have a score here.
- 1 GPT-5.5 88.8%
- 2 GLM-5.2 ↑2 88%
- 3
MiniMax-M3 84.9%
- 4
DeepSeek-V4-Pro ↓2 81.2%
- 5
Qwen3.5-Plus ↑2 79.9%
- 6
DeepSeek-V4-Flash ↓1 78%
- 7
MiniMax-M2.7 ↓1 76.1%
Dynamic behavior
7 models we track have a score here.
- 1 GPT-5.5 65.4%
- 2 GLM-5.2 ↑2 61.9%
- 3
DeepSeek-V4-Pro ↓1 57.4%
- 4
DeepSeek-V4-Flash ↑1 54.1%
- 5
MiniMax-M3 ↓2 52.1%
- 6
Qwen3.5-Plus ↑1 43.1%
- 7
MiniMax-M2.7 ↓1 31.3%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The SemaPLC Project-Context Track paper
GPT-5.5 on the official leaderboard passed 85% in April 2026
Models released before it came out already scored that high. All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SemaPLC Project-Context Track leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Engineering and CAD
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge.
Gemini 4 Carbon 83.9%
- 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity.
LongCat-2.5 55.4%
- CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
- Razavi-Bench Not in model cards Models explain what happens in analog circuits and why. A language model awards up to four points per answer against a reference and a rubric.
Gemini 4 Carbon 98.3%
- CircuitRubric Not in model cards Models write analog circuit netlists from design requests. A program compares transistor connections and relative device sizes with accepted reference circuits. Claude Opus 4.8 96%
Data audit
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (7 entries).
- Pass Models scored 7 models we track.
- Pass Newest score The newest is from June 2026, for GLM-5.2.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5.5 (#21).
- Note Headroom Saturated. The best is 100%, past 85% since April 2026.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 28 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.